Investigating factors of metabolic bone disease in baboons (Papio spp.) using museum collections

1 INTRODUCTION

Metabolic bone disease (MBD) encompasses a range of pathological conditions of bone formation, remodeling, and/or mineralization that can include rickets, osteomalacia, and fibrous osteodystrophy (Canington & Hunt, 2016; Olson et al., 2015; Uhl, 2018; Wharton & Bishop, 2003). This group of conditions is characterized by changes in bioavailable mineral quantities (e.g., calcium or phosphorus), deficiencies in nutrients such as vitamin D, and/or organ dysfunction (Craig et al., 2016; Uhl, 2018; Wolfensohn, 2003). MBD is found across terrestrial vertebrates (Craig et al., 2016), including humans and nonhuman primates (NHPs) (Farrell et al., 2015; Hatt & Sainsbury, 1998; Rajakumar, 2003; Wharton & Bishop, 2003; Wolfensohn, 2003), other mammals (Chesney & Hedberg, 2010; Uhl, 2018), and amphibians and reptiles (Klaphake, 2010). Diet, sunlight exposure, stressful environments, renal disease, genetic factors, and hormonal abnormalities have all been implicated in multifactorial MBD etiology (Farrell et al., 2015; Hannan et al., 2019; Uhl, 2018). Though speculative, gastrointestinal parasites may also play a role in MBD etiology, as noted in in camels (Lynch et al., 1999) and reptiles (Loukopoulos et al., 2007). Osteological markers of MBD include gross bone structural changes, reduced mineral density, and poorly formed cortices and trabeculae, all of which can be regionally localized or found across the skeleton (Adkesson & Langan, 2007; Farrell et al., 2015).

The general diagnosis of “metabolic bone disease” is commonly used instead of specific diagnoses (such as rickets, osteomalacia, or fibrous osteodystrophy) or when multiple conditions are present in an individual (Uhl, 2018). Of all MBDs, fibrous osteodystrophy appears most prominently in the cranium and mandible, whereas rickets and osteomalacia are better detected through postcranial elements (Table 1). The bilateral enlargement of fibrous osteodystrophy lesions in the cranium and mandible is a direct result of osteoclastic bone resorption; bone mineral (calcium and phosphorus) is transferred back to the circulatory system, and distortion of the remaining cancellous bone occurs (Canington & Hunt, 2016; Olson et al., 2015). Rickets represents a softening of the bones linked to vitamin D deficiency during development, which occurs prior to epiphyseal fusion (called osteomalacia during adulthood after epiphyseal fusion has ceased; Wharton & Bishop, 2003).

TABLE 1. Definitions of specific MBD conditions that may be evident in the study sample, following Canington and Hunt (), Craig et al. (), Fiennes (), Farrell et al. (), and Uhl () MBD condition Definition Rickets Developmental bone disease related to vitamin D and/or phosphorus deficiencies and characterized by poor endochondral ossification at epiphyseal growth plates and poor bone formation, often warping bone structure where there is osteoid but no bone mineral formation; similar to osteomalacia but only affects young animals; if persistent in growth, permanent skeletal changes carry on into adulthood Osteomalacia Bone disease related to vitamin D and/or phosphorus deficiencies and changes in modeling and remodeling, where poor bone formation causes lesions similar to rickets and osteoid forms but not bone mineral; does not affect any cartilaginous growth plates because this disease is found in adults only Fibrous Osteodystrophy Lesions formed by bone resorption and replacement with immature bone that lacks proper structure and mineralization, results from primary or secondary hyperparathyroidism; secondary hyperparathyroidism is more common and is due to renal disease, elevated parathyroid hormone, and/or calcium and phosphorus mineral imbalance; bilateral enlargement of the cranium and mandible are especially noted with this condition

While genetic factors may play a role in MBD etiology, nongenetic factors are most often implicated (Farrell et al., 2015; Hannan et al., 2019; Uhl, 2018). MBD presence in animals can be particularly revealing about their environmental conditions during life, including anthropogenic (human constructed or influenced) spaces. Many animals are not equipped to handle the novel, rapidly imposed pressures of some anthropogenic environments, which can alter diet, mobility, social behavior, and reproduction (e.g., primates and carnivores; Michalski & Peres, 2005). MBDs are rarely recorded in non-captive (i.e., “wild”) populations, although nutritionally related MBD has been reported in wild birds, possibly due to a calcium-deficient diet related to range expansion into suboptimal anthropogenic habitats (Cousquer et al., 2007; Phalen et al., 2005). MBD is well recognized in many domesticated animals (including sheep, goats, llamas, alpacas, cattle, pigs, horses, reptiles, cats, and dogs; Dittmer & Thompson, 2011), frequently linked to deficiencies in nutrition and/or sunlight exposure (Dittmer & Thompson, 2011; Uhl, 2018). For animals in captive settings, including zoological parks and research facilities, MBD has long been associated with nutritional and environmental stress (Fiennes, 1974; Mann, 1930; Ratcliffe, 1966; Wackernagel, 1966).

Historical reports of MBD are confounded by a lack of pathophysiological knowledge and consensus on diagnosing zoo animal pathologies. Instead, all MBDs were broadly classified as “rickets.” At the London Zoo in 1890, surgeon John Bland-Sutton noted that “half the monkeys and lemurs brought to this country die rickety” (Bland-Sutton, 1895:266). However, the description of “rickets” in the historical sense more broadly encompassed a suite of MBDs, rather than the specific diagnosis used today. Some historical cases, like “rickets” in lion cubs at the London Zoo in 1889, were most likely fibrous osteodystrophy, where lesions form by bone resorption and replacement with immature bone that lacks proper structure and mineralization (Table 1; Chesney & Hedberg, 2010).

While misdiagnoses were common, historical reports also highlight efforts to change zoological practices for disease prevention; in particular, changes to diets, enclosures, and other animal welfare strategies helped practitioners mitigate MBD prevalence. The first successful treatment of MBD occurred in the aforementioned lion cubs at the London Zoo, after doctors prescribed a diet of goat meat, crushed bones, and cod liver oil. This diet is now known to have a high nutrient content of calcium, phosphorus, vitamin A, and vitamin D, which reversed MBD in these animals (Chesney & Hedberg, 2010). By the 1930s, diet and sunlight had been identified as the key to disease prevention in captive settings (Fiennes, 1974; Mann, 1930; Ratcliffe, 1966; Wackernagel, 1966). As enclosures shifted from restrictive cages to naturalistic designs, zoo veterinarians began to remedy the high prevalence and severity of MBD through dietary means such as vitamin D, calcium, and phosphorus supplements, as well as UV light therapy to produce cutaneous vitamin D (Fiennes, 1974; Ratcliffe, 1966; Wackernagel, 1966). Since animal welfare legislation was enacted in the mid-20th century (Hosey et al., 2009), and alongside greatly improved diets, habitats, and veterinary care practices (Fiennes, 1974; Gutierrez et al., 2021; Smithsonian Institution, 1942, 1952, 1957), MBDs have been greatly reduced in captive NHPs, save a few isolated cases (Hatt & Sainsbury, 1998; Morrisey et al., 1995; Wolfensohn, 2003).

Modern studies of MBDs in NHPs have led to a better understanding of specific MBD pathways and attributes and highlighted the importance of improved captive conditions. Farrell et al. (2015) compared a large sample of captive NHP crania from historical (late-19th to mid-20th century) and recent (1980s and onwards) collections and observed the absence of osteomalacia and other forms of MBD with the modernization of captive NHP care. Baboons showed one of the highest frequencies of MBD among the 12 taxa studied, for example, 8.2% in Papio versus 3.5% in the total sample (Farrell et al., 2015), although the “natural baseline” of MBD in non-captive populations is unknown. Notably, baboons display strong interspecies variability in skin and pelage morphologies, including pigmentation (Hill, 1967; Kamilar, 2006). In both non-captive and captive living baboons, exposed skin color influences vitamin D3 production, but not downstream metabolism (Ziegler et al., 2018). Because of the known connection between sunlight and MBD, it is plausible that skin color differences could lead to variable disease outcomes, as documented in humans (Holick, 2004). In addition, the relative availability of dietary vitamins and minerals, and/or the functionality of their physiological pathways, may also contribute to variable disease outcomes. As far as is known, all catarrhine primates, including Papio baboons, have similar requirements for vitamins and minerals (Milton, 2003), including calcium and phosphorus, and similar levels of circulating parathyroid hormone (Fincham et al., 1993), all of which can cause MBD if imbalances or disruption to physiological pathways occur. There are some physiological differences within the Order Primates specifically related to vitamin D, for example, differential synthesis rates between vitamin D2 and vitamin D3 and different levels of circulating vitamin D (Crissey et al., 1999; Gacad et al., 1992; Marx et al., 1989). These differences are most pronounced between platyrrhine and the vitamin D3-resistant catarrhine primates, the latter having very low vitamin D uptake by target cells (i.e., cell “resistance”) resulting in higher levels of circulating bioactive vitamin D (Adams et al., 1985; Gacad et al., 1992).

In this study, we investigated cranial and mandibular indicators of MBD in baboons (Papio spp.) to better understand the potential health impacts of anthropogenic environments on NHPs. Past studies of MBD have predominantly focused on pathological analysis and approaches to remedying animal disease (e.g., Fiennes, 1974; Ratcliffe, 1966; Wackernagel, 1966). However, we take an ecological approach to the study of MBD, considering how diseases can arise when animals are subjected to environmental conditions that differ from their natural habitat (Eller et al., 2019). Using a large sample of non-captive and captive baboon cranial and mandibular specimens spanning nearly a century (1890–1971) at the Smithsonian Institution's National Museum of Natural History (NMNH), we expand on work by Farrell et al. (2015) and Ziegler et al. (2018) by examining environmental and physiological factors and skeletal manifestations of MBD. We hypothesize that anthropogenic environments are conducive to MBD in NHPs. Accordingly, we test three predictions: 1) the captive group has a higher MBD frequency than the non-captive group, 2) the difference in MBD frequency between captive and non-captive groups decreases as captive conditions improve over time, and 3) skin color groups do not differ in MBD frequency, where baboons with darker facial skin color do not have a higher incidence of MBD than baboons with lighter facial skin color. This final prediction addresses the complexity in MBD etiology. Both skin color (as a proxy for cutaneous vitamin D synthesis; Ziegler et al., 2018) and diet (vitamins and minerals) influence MBD etiology, although the latter is not tested here.

2 MATERIALS AND METHODS 2.1 Sample

We macroscopically examined 160 baboon (Papio spp.) skulls in the Division of Mammals at the NMNH (Supplementary Table 1). Due to the over-representation of primate skulls in the NMNH collections relative to postcranial elements, we focused on cranial and mandibular manifestations of MBD to allow for a larger sample size. Table 2 summarizes the demographic information collected from specimen tags and NMNH accession records for each specimen. Taxonomic designations followed the Zinner et al. (2011) scheme, which include six distinct Papio species: P. anubis, P. papio, P. ursinus, P. cynocephalus, P. kindae, and P. hamadryas.

TABLE 2. Taxonomic distribution of specimens by skin color group, sex, age, accession year, and environment type Taxon Total Color Sex Age Accession year Environment Female Male Infant Juvenile Subadult Adult 1890–1949 1950–1971 Captive Wild P. h. anubis 69 D 31 38 8 6 9 46 29 40 45 24 P. h. cynocephalus 24 M 14 10 1 3 5 15 2 22 19 5 P. h. hamadryas 9 L 3 6 1 3 2 3 7 2 9 0 P. h. papio 17 D 2 15 0 6 1 10 2 15 11 6 P. h. ursinus 41 M 17 24 2 4 7 28 24 17 4 37 Total 160 n/a 67 93 12 22 24 102 64 96 88 72 Note: Specimens were classified as “captive” if the individual lived in a captivity at any point during life, regardless of birthplace. Accession years were grouped for descriptive purposes but not for statistical testing.

Environmental assignments were based on whether an individual died in their native African habitat (“non-captive”) or in captivity (“captive”), including zoological parks and biomedical facilities. Non-captive individuals were further geographically and temporally subdivided to consider habitat diversity across Africa (Supplementary Table 2). Additionally, the specimen's date of accession (the most reliable date available) was recorded from each tag, indicating the year that NMNH acquired the remains, usually soon after death (Supplementary Table 3).

Age was estimated using molar eruption stages following Kahumbu and Eley (1991) and Phillips-Conroy and Jolly (1988). Each specimen was assigned to one of four categories: infant (no molars fully erupted, <20 months old), juvenile (all first molars completely erupted, 20–49 months old), subadult (all second molars completely erupted, 50–80 months old), or adult (all third molars completely erupted, >80 months old). Molar eruption is more strongly correlated with age than are other tooth eruption sequences or cranial characteristics (e.g., sutural fusion and cranial size), all of which may confound aging estimates because of their relationship with sex and social dominance rank (Galbany et al., 2015; Phillips-Conroy & Jolly, 1988). For the purposes of this study, subadult describes an individual approaching reproductive maturation. Sex was primarily determined from the specimen tag but also assessed by canine size, a highly sexually dimorphic feature in adult baboons (Phillips-Conroy & Jolly, 1981).

Facial skin pigmentation was classified into three color groups: light, medium, and dark (Figure 1). Following Ziegler et al. (2018), three baboon species were categorized in their visually determined color scale: P. anubis as dark, P. cynocephalus and P. kindae as medium, and P. hamadryas as light. For P. papio and P. ursinus, we used Kingdon et al. (2013) for skin and pelage pigmentation descriptions to place P. papio in the “dark” category and P. ursinus in the “medium” category. When possible, associated specimen skins were visually assessed to confirm the group assignment.

image

Skin color categories used to assess the influence of skin color on cranial and mandibular indicators of MBD. (Image credits: Pixabay [left]; Alexander Landfair, University of Dar es Salaam via Wikipedia [middle]; Sharp Photography via Wikimedia Commons [right])

2.2 MBD assessment

Published photos and gross descriptions of MBD cranial indicators in humans and nonhuman primates and domesticated animals provided the basis for our skeletal assessments (Canington & Hunt, 2016; Craig et al., 2016; Farrell et al., 2015; Fiennes, 1974; Uhl, 2018). Because of the longstanding confusion on specific MBD conditions, modern definitions relevant to the current study are provided in Table 1, following Craig et al. (2016). Although MBD does present postcranially, there are far fewer postcranial skeletons available for study at the NMNH (n = 9), and thus only cranial and mandibular evidence was considered. As MBD broadly encompasses a wide range of conditions, we identified a series of pathological criteria, any of which can be indicative of MBD by visual assessment (Figure 2; Table 3). MBD pathological designations were made by agreement of all authors.

image

Baboon specimens in the study sample that exhibit variation with respect to MBD criteria, from complete absence to multiple indicators consistent with fibrous osteodystrophy among other possible MBDs. (a) non-captive P. anubis individual (USNM 216605) with no cranial evidence of MBD. (b) Captive P. anubis individual (USNM 83980) with minor MBD pathologies including (1) enamel hypoplasia of the incisors and canines, (2) very slight enlargement of the maxilla, and (2) porosity along the maxilla, supraorbital tori, and glabella. (c) Captive P. ursinus individual (USNM 252089) with extreme MBD pathologies including (3) extreme porosity, pitting, and enlargement of the neurocranium and (4) maxilla with medially oriented incisors and canines. (Photo credit: Lucia RM Martino, Smithsonian Institution)

TABLE 3. Pathological criteria used to identify MBD in the study sample MDB criteria References Observed in sample (n) Bone porosity in any of the following skull regions: sagittal suture, glabella, and maxilla often associated with pitting; mandible associated with swollen and spongy look; general increase in bone porosity Farrell et al., 2015 35 Bone thickening and enlargement in any of the following skull regions: frontal, parietal, and/or temporal bones of the cranium; zygomatic bones extending anteriorly into maxilla; maxilla with bilateral enlargement from alveolar margins to frontal process; mandible; nasal conchae; and palate Canington & Hunt, 2016; Farrell et al., 2015; Fiennes, 1974; Uhl, 2018 35 Mandibular condyles are underdeveloped and/or coronoid processes are missing Canington & Hunt, 2016; Farrell et al., 2015 10 Rounded orbital margins with softened orbital edge Farrell et al., 2015 8 Eruption and position of teeth affected when maxilla and mandible has swelled Farrell et al., 2015; Fiennes, 1974 4 General decrease in bone density Canington & Hunt, 2016; Farrell et al., 2015 3 Linear enamel hypoplasia in teeth Guatelli-Steinberg and Skinner, 2000 4 Fragile, friable, and/or brittle bone regions with inner trabeculae exposure in some cases Canington & Hunt, 2016; Farrell et al., 2015 2 Lesions of the skull Uhl, 2018 1 Widening and distortion of cranial and/or facial sutures, almost disarticulated in some cases Farrell et al., 2015 0 Bowing of mandibular rami Canington & Hunt, 2016 0 2.3 Statistical analysis

We used binomial-family generalized linear model (GLMs) to assess differences in MBD frequency by environment (captive, non-captive), specimen accession year, and skin color. The binomial-family GLM uses logistic regression analysis to predict a binary outcome based on explanatory variable(s); the binary outcome in this analysis is presence or absence of MBD. For the specimen accession year analysis, we built two additional binomial-family GLMs, one for the captive specimen subsample (n = 88) and one for the non-captive specimen subsample (n = 72), to assess temporal changes in MBD frequency within each environmental condition. Age, sex, and species were also included as covariates in the five GLMs, examining MBD frequency by: environment, specimen accession year, and skin color. Finally, environment was included as a covariate in the accession year and skin color GLMs. All analyses were conducted with the R statistical programming language (R Core Team, 2021).

3 RESULTS

Most of the captive baboons in our sample lived and died in the Smithsonian's National Zoo and Conservation Biology Institute (NZP) in Washington, DC (n = 26; 29.5%) and the Southwest Foundation for Research and Education in San Antonio, Texas (n = 61; 69.3%). One specimen came from the Oregon Regional Primate Center (Beaverton, OR) and one from the Barnum and Bailey Circus. The non-captive baboons in our sample are predominantly from southern and eastern African localities (n = 62; 86.1% of non-captive baboons) while the remainder are from central and western Africa (n = 10; Supplementary Table 2). Skeletal indicators of MBD for each specimen are reported in Supplementary Table 4. Of the 160 specimens, 51 (31.9%) showed evidence for MBD pathologies, with the highest pathological prevalence related to bone thickening, enlargement, and porosity of different regions of the skull (Figure 2; Table 3).

Of the 51 pathological individuals, 14 died in their native habitats (killed during hunting and specimen collection trips), while 37 died in captivity. For three individuals, the environment is unknown. Within the total sample (N = 160), captive and non-captive specimens differed significantly in MBD frequency (LRT: χ2 = 8.0, p < 0.01) with captive specimens showing a higher frequency of MBD (42%; Figure 3a) than non-captive specimens (19%; Figure 3b). MBD frequency decreased significantly among captive baboons over time (1890–1971; LRT: χ2 = 6.5, p = 0.01; Figure 3a). In non-captive baboons, there was no significant change in MBD frequency over time (1890–1971; LRT: χ2 = 2.2, p = 0.14; Figure 3b). The 51 specimens with MBD pathologies represent five of the six species, excluding P. kindae, and all three categories of exposed facial skin pigmentation (Figure 1). Facial skin color did not significantly predict MBD frequency (LRT: χ2 = 5.6, p = 0.06). MBD was not successively more frequent among the “medium” and “dark” groups as compared to the “light” group (Table 4). In the three statistical models examining MBD frequency by environment, specimen accession year, and skin color, age was a significant predictor of MBD (χ2 = 10.3–12.9, p < 0.05; Table 5), while sex (Table 6) and species designation were not.

image

Temporal change in MBD frequency by decade from 1890–1971. Frequencies are expressed as percentages, and there was a high variance in sample sizes across decade bins. (a) MBD frequency in captive baboons decreases over time. (b) MBD frequency in non-captive baboons does not change over time. Blank decades do not have any represented specimens

TABLE 4. Distribution of MBD pathologies between the skin color categories, including light, medium, and dark categories Color MBD No MBD Total % MBD Light 7 2 9 77.8 Medium 20 45 65 30.8 Dark 24 62 86 27.9 Total 51 109 160 31.9 TABLE 5. Age distribution of specimens by environment Captive Non-captive Total % MBD Total % MBD Infant 5 60.0 7 28.6 Juvenile 16 81.3 6 16.7 Subadult 9 66.7 15 13.3 Adult 58 25.9 44 20.5 All 88 42.0 72 19.4 Note: Frequency of MBD is noted for each age group. TABLE 6. Sex distribution of specimens by environment Captive Non-captive Total % MBD Total % MBD Female 37 29.7 30 16.7 Male 51 51.0

留言 (0)

沒有登入
gif